👁️ ZenLens: OCR & Website Translator

ZenLens is a high-performance, aesthetically pleasing web application designed to capture, recognize, and translate text from any website or image. Built with Next.js 16 and React 19, it combines browser-native OCR with local AI vision capabilities.

(Placeholder for your banner)

✨ Features

• 🚀 Hybrid OCR Engine:
  • Tesseract.js: Fast, browser-side text recognition for standard tasks.
  • Aether Vision (Ollama): High-fidelity OCR using local AI models (Llava, Bakllava) for complex layouts.
• 🌍 Instant Translation: seamlessly translate English text to Turkish (and more) using the MyMemory API.
• 📸 Intelligent Capture: Built-in logic to capture specific areas of a website or uploaded images.
• 🎨 Glassmorphic UI: A premium, dark-mode interface with smooth animations powered by Framer Motion.
• ⚡ Local-First: Your data stays on your machine when using the Ollama engine.

🛠️ Tech Stack

• Framework: Next.js 16 (App Router)
• Library: React 19
• Styling: Vanilla CSS with CSS Variables (Glassmorphism)
• Icons: Lucide React
• Animations: Framer Motion
• OCR: Tesseract.js & Ollama API
• Language: TypeScript

🚀 Getting Started

Prerequisites

• Node.js 18+
• Ollama (required for Aether AI OCR)
  • Run the following to prepare the vision model:

    ollama run llava

Environment Variables

If you want to use external translation/OCR engines without entering keys every time, you can set them in a .env.local file:

OPENROUTER_API_KEY=your_key_here
OCR_API_KEY=your_key_here
NEXT_PUBLIC_OLLAMA_URL=http://localhost:11434

See .env.example for details.

Installation

1. Clone the repository:

   git clone https://github.com/halilogia/ocr-translate-websites.git
   cd ocr-translate-websites

2. Install dependencies:

   npm install

3. Run the development server:

   npm run dev

4. Visit the app: Open http://localhost:3000 in your browser.

📖 Usage

1. OCR Scanning: Select "OCR Scanner" from the sidebar. Upload an image or use the capture tool to select text.
2. AI Vision: Toggle the "Aether" engine to use your local Ollama instance for better accuracy in complex images.
3. Translation: Recognized text is automatically synced to the Translator, providing instant Turkish equivalents.

---

📄 License

Distributed under the MIT License. See LICENSE for more information.

---

Developed with ❤️ by Halilogia